Journal of Industrial Ecology最新文献

Rapid advancements in artificial intelligence (AI) are driving transformative changes in many areas, with significant environmental implications. Yet, environmental assessments for specific applications are scarce. This study presents an in-depth life cycle assessment of “Foodforecast,” a machine learning (ML) cloud service designed to reduce food waste in bakeries by optimizing sales forecasting. It covers four impact categories: global warming, abiotic resource depletion, cumulative energy demand, and freshwater eutrophication. The assessment includes both the direct environmental impacts of the ML model and the underlying system hardware, as well as the indirect benefits of avoided bakery returns compared to traditional ordering methods, using real-world case study data. In 2022, “Foodforecast” led to an average 30% reduction in bakery returns, primarily bread and rolls, according to sales reports. The associated environmental benefits significantly outweighed the system's direct impacts by one to three orders of magnitude across impact categories and return utilization scenarios. The study identifies support activities such as service maintenance during deployment as major direct impact factors, surpassing those from cloud compute for ML operations. Data processing and inference dominate the latter, while the much-discussed ML training plays a minor role. The environmental consequences of AI are complex and dual sided. This case study demonstrates that AI might provide environmental benefits in certain contexts, yet results are constrained by methodological challenges and data uncertainties. There remains a need for further holistic LCAs across different ML applications to inform decision-making processes and ultimately guide the responsible use of AI.

Important expectations are placed globally on the private sector to take part in co-governing sustainability challenges. With increasing recognition that business organizations depend and impact on natural capital, biodiversity, and related ecosystem services, it has become pivotal for companies to be able to appraise and manage such issues. This calls for developing avenues through which biodiversity and ecosystem services can be incorporated in sustainability accounting (as well as reporting and response practices) by individual organizations, without neglecting a value chain-wide perspective, and aligning that with existing efforts for public accounting of natural capital, against the overall framework of national and global sustainability goal and targets. This article addresses such a call by elaborating on the contributions of ecosystem services to business organizations and illustrating the two main challenges related to feeding private sector data into national accounting of natural capital and ecosystem services; and to understand how the two challenges identified in the previous point can be addressed by companies, we provide an overview of the fragmented landscape of management systems, approaches, methods, and initiatives dedicated to monitoring, reporting on, and responding to biodiversity and ecosystem services issues at the organizational and value chain level. We conclude by offering reflections on how to foster a shift from one-off assessments at the company level to more systematic and comprehensive ones along the value chain.

Greenhouse gas (GHG) emissions datasets are often incomplete due to inconsistent reporting and poor transparency. Filling the gaps in these datasets allows for more accurate targeting of strategies aiming to accelerate the reduction of GHG emissions. This study evaluates the potential of machine learning methods to automate the completion of GHG datasets. We use three datasets of increasing complexity with 18 different gap-filling methods and provide a guide to which methods are useful in which circumstances. If few dataset features are available, or the gap consists only of a missing time step in a record, then simple interpolation is often the most accurate method and complex models should be avoided. However, if more features are available and the gap involves non-reporting emitters, then machine learning methods can be more accurate than simple extrapolation. Furthermore, the secondary output of feature importance from complex models allows for data collection prioritization to accelerate the improvement of datasets. Graph-based methods are particularly scalable due to the ease of updating predictions given new data and incorporating multimodal data sources. This study can serve as a guide to the community upon which to base ever more integrated frameworks for automated detailed GHG emissions estimations, and implementation guidance is available at https://hackmd.io/@luke-scot/ML-for-GHG-database-completion and https://doi.org/10.5281/zenodo.10463104. This article met the requirements for a gold-gold JIE data openness badge described at http://jie.click/badges.

In industrial ecology, approaches have been developed to analyze the contribution of specific sectors to environmental impacts within supply chains. In economics, a range of methods addresses the forward linkage (use of output) and backward linkage (dependency on inputs) of sectors, and the analysis of key sectors. This article offers a formal investigation of the relationship between these. It shows that both the analysis of supply chain impacts and of intersectoral linkages can be seen as special cases of a more general hypothetical extraction method (HEM). In HEM, sectors' role is assessed as the effect of their removal on the input–output model's solution. HEM also allows for the (partial) extraction of individual transactions. HEM thus offers a flexible approach to assessing the contribution of one or several sectors, or transactions, or parts thereof, to value added or footprint of any final demand. It can be applied to study the environmental footprints of companies or intermediate products, the contribution of certain inputs to sectors, or the potential impact of disruptions of supply chains on producers and consumers. In this article, the price model for HEM is introduced to identify the contribution of the extracted (target) sectors to the price or unit footprint of a commodity.

The low-carbon energy transition requires significant increases in production for many mineral commodities. Understanding demand, technological requirements, and prices associated with this production increase requires understanding the supply chain dynamics of many minerals simultaneously, and via a consistent framework. A generalized economics-informed material flow method, global materials modeling using Bayesian optimization, captures the market dynamics of key mineral commodities. The method relies only on a limited set of widely available historical data as input, enabling quantification of economic relationships (elasticities) for supply chain components where data are sparse, and relationships cannot be obtained via traditional statistical approaches. Building upon established material flow analysis (MFA) and economic modeling techniques, Bayesian optimization was applied to fit an economics-informed MFA model to global historical demand, supply, and price for aluminum, copper, gold, lead, nickel, silver, iron, tin, and zinc. This approach enables estimates for the evolution of ore grades, mine costs, refining charges, sector-specific demand, and scrap collection for each commodity. Economic relationships were quantified and compared with a database compiled from the literature, including 1333 values from 213 analyses across 65 publications. Discrepancies in methods and limited coverage make use of these parameters in modeling efforts difficult. This work provides a single, homogeneous, probabilistic approach to identifying economic relationships across mineral supply chains, with uncertainty quantification, a literature database for comparison, and a modeling framework in which to use them. This article met the requirements for a Gold-Gold JIE data openness badge described at http://jie.click/badges.

In this paper, we take an axiomatic approach to the design of ecological footprint indices. Our focus is put on the heterogeneity of land with respect to types and regions, at the core of an inherent aggregation problem. We propose an axiomatic characterization of the ecological footprint index with two fundamentally new axioms, symmetry and independence, which can resolve the problem of land heterogeneity. It is shown that a unique index, up to an affine transformation, exists meeting the axiom system. This index simplifies the aggregation procedure considerably and avoids the need for a synthetic unit of measurement, like global hectares, as well as complex transformations of variables by means of weighting schemes. Our findings reveal differences with the Global Footprint Network (GFN) index, in particular with regard to the treatment of land heterogeneity. Finally, the axiomatic methodology employed may open up perspectives for the development of ecological measures in general, and especially of measures for sustainability and tipping points.

Emergy analysis (EmA) is the quantitative sustainability method that analyses materials, energy resources, goods, or services under a common unit of solar emergy (seJ). Meanwhile, life cycle assessment (LCA) is one of the popular tools to evaluate environmental impacts. Both methods have been widely used in various applications, hence, the idea of co-jointing these two methods has been increasing to optimize the assessment's inclusivity. However, the existing integrated LCA–EmA methods are applied segmentally with minimal data integration, which has limited the potential of co-benefits in sustainability accounting. This research proposes a more comprehensive integration between LCA and EmA, known as life cycle emergy analysis (LCEmA). The process data is fully integrated, and the detailed methodological approach is presented to demonstrate the assessment process. Diaper manufacturing is selected as a case study to validate the functionality of the proposed LCEmA approach. The inclusion of foreground and background data offered in LCEmA approach considers backend processes such as electricity generation for manufacturing activities. This has shifted the sustainability hotspot from raw material extraction to the manufacturing process. This research demonstrates that the LCEmA approach can perform comprehensive analysis as a promising alternative to the existing integrated LCA–EmA methods.

The assessment of energy consumption of data traffic for Internet services usually relies on energy intensity figures (in Wh/GB). In this paper, we argue against using these indicators for evaluating the evolution of energy consumption of data transmission induced by changes in Internet usage. We describe a model that estimates global impacts for different scenarios of Internet usages and technological hypothesises, and show that it can overcome some limitations of intensity indicators. We experiment the model on four use-cases: basic usage, video streaming, large downloads, and video conferencing. Results show that increasing the resolution of videos does increase the total energy consumption while misleadingly decreasing the power intensity indicator at the same time. In other words, a more efficient network does not necessarily mean less energy consumption.

The difficulty in identifying appropriate metrics for the impacts organizations have on biodiversity remains a major barrier to the inclusion of biodiversity considerations in environmentally responsible investment. We derive and analyze a simple metric: the sum of the proportional changes in the global abundances of species caused by an organization, with a regularization to cover the case of species close to extinction. We argue mathematically that this metric quantifies changes in the mean long-term global survival probability of species. The metric thus supports the objective “to significantly reduce extinction risk” of the 2022 Global Biodiversity Framework and allows organizations to set themselves corresponding science-based targets. We show that in a perfect market trade in biodiversity credits quantified by our metric would lead to near-optimal allocation of resources to species conservation. We further show that metric values are quantitatively convertible to several other metrics and indices already in use. Barriers to adoption are therefore low. Used in conjunction with metrics addressing ecosystem services and integrity, potential areas of application include biodiversity related financial disclosures and voluntary or legislated biodiversity credit or no net loss policies.

In this study we quantify wood flows from raw materials to end-uses for the United Kingdom in a robust way using a new material flow analysis (MFA) model with uncertainty. This is important to identify opportunities for efficiency given constraints on wood supply. We have developed a new “ProbPACTOT” MFA model which introduces a systematic method to handle mixed units during reconciliation and uncertainty in data observations, conversion factors, and process recipes. This makes it possible to track material flows all the way to end-uses, which is otherwise difficult because the diverse materials, data types, and units used to quantify end-products are hard to integrate into standard allocation-based MFA models. We apply the model for the case of wood in the United Kingdom by defining 56 process recipes and reconciling 117 data observations from various sources. The results quantify upstream production and trade flows, through to 19 specific end-uses of wood fibers. We use this to show the potential scale of savings by enhancing material efficiency; for example, if pallets were used 25% more intensively, 0.49 $�\pm$0.2 $��M�m^3�$ of wood fibers could be saved, corresponding to 4%–7% of the total soft sawnwood consumption of the United Kingdom. Judging the scale of opportunities for wood material efficiency in the United Kingdom is important domestically, and has global significance as the United Kingdom is the second largest net importer of wood products in the world. Moreover, this study proposes an important advancement in MFA giving a structure for modeling uncertain material flows up to end-uses, applicable to any material. This article met the requirements for a gold-gold JIE data openness badge described at http://jie.click/badges.